Image 01 Image 03

Ten Polls

Ten Polls

Reader Charles has been a great source of polling information, so I asked him to put together a list of the 10 most interesting recent polls in Senate and House races. 

Here is his list, not in any particular order:

GA-12 (Barrow – D) ( Cook Index R+9 ) (this is an Operation Counterweight race)
D Poll 10/12 — 48  45 D+3
Tenn-4 (DesJarlais, R) ( Cook Index R+13)
D Poll 10 /15 — 49 /44 R +5
NV-4 NEW (Cook Index D+2)
R Poll : 10/10  — R+13
NY-25 (Slaughter-D )(Cook Index D+5)(This is Operation Counterweight race)
Siena Poll 10/10 — 49/44 D+ 5
Illinois Districts 8, 10 11, 12, 13 ,17
We Ask America Poll 10/10
Marquette Poll 10/14 — 46/45 R+1
Siena Poll 10/14 — 46 / 44 D +2
Quinnipiac Poll 10 /14 — 48 / 45 D+3
Wenzel Strategies R poll 10 /13 — 49 /45 R+4 
Montana Senate
Rasmussen Poll 10 /14 —  48 /48 Tied  
Some other resources:
Unskewed Presidential Battleground State Polls ( 2004 , 2008 & Blended Party ID )
Ohio Early Voting 2012 vs. 2008

Donations tax deductible
to the full extent allowed by law.


Not sure why these polls qualify as interesting. We should care more about accuracy. For this, we should not look to random readers, but rather to those with proven track records for providing filtered and aggregated info, such as Nate Silver. Anything else is just wishful thinking, and unlikely to lead to an optimal assignment of resources.

    legalizehazing in reply to Avi. | October 17, 2012 at 8:22 pm

    It is interesting to me because I haven’t been following the numbers at all. They may not be too valuable but I didn’t even have a ballpark idea. I appreciate the aggregate picture in one place as well.

Rasmussen reports that Richard Mourdock now holds a five point lead over Democrat Congressman Joe Donnelly in the Indiana Senate race.

Those poll results are after the latest debate between the candidates where Mourdock nailed Donnelly for supporting Obama’s programs.

The results are despite the continued refusal by Dick Lugar to come out for Mourdock in order to move some Lugar moderates to the Republican ledger. Interestingly, Lugar came out in support of Scott Brown – so I guess moderates hang together.

    TrooperJohnSmith in reply to gad-fly. | October 17, 2012 at 10:18 pm

    “Dick Lugar” sounds like a 1950’s pulp magazine private-dick. Like his fictional counterpart, the real man always seemed one-dimensional and stuck in another time, out of touch with middle America. In a word: RINO.

That TN-04 race is only interesting because the incumbent Republican got some phone transcripts dumped on him. Apparently in 2000, he talked his mistress (while he was married) into getting a divorce. Tricky situation.

    NC Mountain Girl in reply to matthewhurtt. | October 18, 2012 at 12:00 am

    It’s worse than that. He supposedly tried to talk his mistress, who was also his patient, into getting an abortion. He was separated from his first wife, says he still wanted to save the marriage and supposedly the mistress wasn’t actually pregnant. He was only trying to get her to admit that by demanding an abortion.

    This rural district sprawls over two dozen counties and across several TV and radio markets. The good news is since the district is on the fringe of most of those markets the story isn’t likely to get a huge amount of media attention. The bad news is other people’s sins tend to be a favorite topic of conversation in rural districts. It will be hard to counter all the embroidery likely to attach to this tale as it spreads.

      Oh, what a tangled web we weave…

      Fascinating tale, NCMtnGirl. We Iowans aren’t above a bit of gossip either but most of our pols are rather uninteresting lumps. Your election is pretty spicy.

NC Mountain Girl | October 17, 2012 at 11:15 pm

I suspect Casey could be toast if he is only at 48% in a Quinnipiac poll because I’ve never looked at a Q poll that wasn’t skewed left. I am also happy to see that Thompson is up but within the MoE in the Marquette poll. Those late primaries can be rough. Thompson’s war chest was depleted, he didn’t have a lot of time to mend fences with the losers and Baldwin was already on the air with negative ads.

I find the Siena poll for the Connecticut Senate race useless. The sample was very small for a statewide race -only 552 likely voters – and yet it was taken over 12 days! A lot happened during that period.

There is no link to internal information for the Siena poll in NY-25. I have seen past Siena polls in Congressional races that were very close to the results on election day. In those polls Siena started with a list of phone numbers for registered voters in the District rather than random digit dialing. They applied a good likely voter screen, had a sufficient sample size and did it all in only three day.

The negatives for both major party candidates in the Missouri Senate race are sky high. I worry that the only other candidate in the race is a Libertarian, which may be bad news for Akin. I was hoping for a Green to take votes from McKaskill.

I wouldn’t give you a nickel for any poll that doesn’t use a likely voter screen (and even then a nickel’s all I’d give you for any poll right now). This election hinges 100% on turnout. A lot of poll respondents will tell you they are likely to vote on election day but if they wake up and it’s raining cats and dogs they may not make it.

As absentee voting becomes more and more prevalent it seems polls ought to include a provision for “already voted”. There is where you’ll get an idea of motivated voters.